Re: Proposal: 'tag' URIs

On Fri, Apr 27, 2001 at 10:04:02AM -0700, Tim Kindberg wrote:
> At 05:27 PM 4/27/2001 +0100, Sean B. Palmer wrote:
> >In particular, tag could register urn:tag: (or whatever) and decree
> >that that space is to be used for what is currently known as "tag:"
> >URIs. In development, one could use "urn:x-tag:". As Michael points
> >out, there is no specific problem with using URNs for this particular
> >application - they are unique, non-reassignable, and they need not
> >resolve to any particular entity. I just don't see why one should use
> >up an entire URI scheme on this when the particular semantics are
> >already covered by an exisiting URI scheme.
> 
> I guess I'm either missing the point or not effectively making my point 
> about the resolution assumptions behind URNs -- location-independent vs 
> contextual -- but in my mind that still eliminates URNs, despite the 
> advantages that using the URN namespace would have.
> 
> The literature on URNs is littered with 'location independent'. So what am 
> I missing, exactly?

I think there's a mismatch on the definition of 'location independent'. 
Location independent in those documents simply mean that there is no
requirement that the identifier by resolved against some specific network
service that also is part of the name. Any identifier can be resolved
contextually since contextual resolution is a function of who you ask
questions about the identifier, not the identifier itself. And since
RFC 2141 doesn't even define resolution you can resolve it based on
any context you trust becase you are the one defining what 'resolution'
means for your application....

-MM





-- 
--------------------------------------------------------------------------------
Michael Mealling	|      Vote Libertarian!       | www.rwhois.net/michael
Sr. Research Engineer   |   www.ga.lp.org/gwinnett     | ICQ#:         14198821
Network Solutions	|          www.lp.org          |  michaelm@netsol.com

Received on Friday, 27 April 2001 14:21:39 UTC